President Clinton's plan to extend Social Security's solvency through 2055 preserves the existing benefit structure and rejects individual accounts. The plan relies primarily on placing most of the projected 15-year budget surplus into the Social Security trust fund. These aspects of the Clinton plan have been warmly welcomed by progressives who feared an imminent compromise that would have added individual accounts by diverting revenue from the current system. Partly to increase the projected rate of return, and partly to steal the privatizers' thunder, the President also proposed placing some $600 billion of the projected surplus into the stock market to earn a higher return. This part of the plan warrants a more wary reception.
Placing some of the trust fund in the stock market raises both phony issues and real ones. Let's dispatch the bogus ones first. For one thing, the amount of money the administration proposes placing in the stock market will never be a very large share of the trust fund or the stock market. At present, the trust fund holds approximately $800 billion in government bonds. Under existing law, this would increase to $3.35 trillion by 2015. With the additional amount that Clinton has proposed to add from general revenue, it should reach $6.25 trillion by 2015. The $600 billion to be placed in the stock market would be less than 10 percent of the system's total assets at that point. And the proposal calls for capping at 14.6 percent the portion of the trust fund held in stock. So only a limited portion of the trust fund would ever be placed at risk.
Currently the value of all publicly traded stock in the United States is approximately $13 trillion. If the stock market grows at the same pace as the economy over the next 15 years, it will rise to approximately $27 trillion by 2015. The $600 billion held by the trust fund would then be less than 2.5 percent of the stock market. This is real money, but far from controlling interest. By itself, that infusion should not significantly affect stock prices, though it might spur the "irrational exuberance" that appears to have been driving the market in recent years. In short, the survival of Social Security will not depend on the stock market, nor will the government take over corporate America through its stock purchases. (At the same time, a significant downturn in the stock market would produce a shortfall in the system's reserves. And the government would hold a large enough stake in the market that it could be a serious factor, should it choose to have a real impact on corporate policy.)
There are those who are concerned that placing money in the stock market will prevent federal dollars from going to public investment or other social needs. It is important to realize that under the Presi dent's proposal, the money placed in the trust fund will be treated in exactly the same manner as the money placed in the stock market. Both will be counted as expenditures.
Nor will putting money in the stock market, either through individual accounts or collectively, affect the national savings rate. There is no economic theory that suggests the economy would grow any faster as a result of having the money placed in the stock market.
Some critics, including Alan Greenspan, have contended that government is a poor picker of stocks, and that the returns from such government investments tend to be inferior to private returns. In fact, the government begins with one big advantage, in that middleman commissions would be far lower than those in private investments. A potential disadvantage is that government would be necessarily more risk-averse; and since risk tends to be correlated with reward (and also sometimes with catastrophic loss), one would expect government holdings to show not only less volatility but also a lower average rate of return. In fact, though the evidence on public pension funds is limited, it does not indicate that they will necessarily have poorer returns than private funds.
For example, a recent study by Olivia Mitchell and Ping-Lung Hsin found that the average return on 128 public pension plans for the period from 1986 to 1990 was 11.6 percent. This was somewhat higher than the 10.1 percent average stock return in this period, but somewhat lower than the 13.5 percent average return on government bonds. In short, this is the sort of yield that would be expected from a fund with a typical mix of stock and bond holdings and low administrative costs. Whatever political involvement or risk aversion existed in the investment decisions of these funds apparently did not have a negative impact on average returns in this period, although in some funds there may have been a significant effect. (Also, in prior years the returns on these funds fell somewhat below market averages.) Studies of private sector pension funds have shown that they often significantly underperform market indices, even before deducting management fees. While additional research may shed more light on the topic, the evidence to date does not provide any reason to believe that if Social Security's assets were invested in the stock market, the return to the trust fund would be any lower than that of private funds.
There are, however, some genuinely thorny issues raised by the Clinton plan:
- the expected rate of return on stock holdings
- the potential loss to Social Security reserves from bear markets
- the questions of corporate governance when the government holds stock
- the political effect of the government having a major stake in private corporations
The main financial motivation for moving some reserves from bonds to stocks is the prospect of higher returns. Clinton's plan assumes that the average real (inflation-adjusted) return on stock will be 6.75 percent a year. This compares to a real return on government bonds projected to be 2.8 percent. The difference in these projected returns extends the solvency of the trust fund from 2049, if all the money were placed in government bonds, to 2055.
Also, the stock projections are based on simple extrapolations from the past. The problem with such extrapolations is that the economy is projected to grow far more slowly in the future than in the past. This is why the Social Security system is projected to show a shortfall. (If future economic growth reflected past growth rates, most of the Social Security shortfall would disappear.) Also, record high price-to-earnings ratios have pushed dividend yields (in percentage terms) to record lows, since dividend payouts have not kept pace with stock prices. In short, the high returns projected for stock are inconsistent with the rest of the projections in the Social Security Trustees' report.
If the economy and profits grow as projected, at just a 1.5 percent annual rate, and with dividend yields currently under 2 percent, the real returns on stock according to my calculations will be in the neighborhood of 3.5 percent, not the 6.75 percent assumed by the President. With $600 billion in stock purchases, the difference between a 3.5 percent stock yield and a 2.8 percent bond yield will barely be enough to extend the life of the program for a year. This more realistic assessment of prospective returns is likely to remove much of the enthusiasm for investing the trust fund in the stock market. Even if we assume a more robust growth rate of 2.5 percent, the difference between the bond yield and the stock yield would only extend the solvency horizon by less than two years.
There is also a real political risk in using overly sunny projections for stock market returns. The main selling point for privatization is that workers could get rich if they were allowed to invest their money themselves. But this compares apples and oranges. Workers who put their payroll tax money directly into the stock market would lose all of the advantages of social insurance, and they would face the risk of down markets as individuals. Defenders of Social Security surrender a key argument by taking the stock projections of the privatizers at face value.
Social Security, unlike individual retirement accounts, is guaranteed by the government. But with government investment in the stock market, government is at risk for sudden drops in the value of its stock portfolio. When the trust fund starts selling off its assets in order to finance its payouts around 2020, its sale of stock will be counted as revenue for the government (the cashing in of government bonds is not counted as revenue.) A certain revenue stream would be projected in these years based on a projected path of stock sales. If the stock market slumped suddenly, the government would be forced either to sell off stock at depressed prices or to cut spending or raise taxes to make up the shortfall. Of course, the rational way to deal with this situation would be to simply have the government borrow to finance its revenue shortfall and hold onto its stocks until the market recovered. But if balanced budget orthodoxy holds sway, government would have to take losses from forced sales of stock in a down market.
For example, the projections show that in 2040 the trust fund would have to sell off $1.147 trillion in assets to pay all its bills. If 14.5 percent of the fund is in stocks, this would mean selling $166.3 billion of stocks in that year. If the stock market had recently taken a plunge, then this amount of revenue would have to be made up from other sources in order to avoid selling at a loss.
Before we panic, however, note that GDP is projected to be $61.621 trillion in 2040. The shortfall due to not selling the trust fund's stocks would be 0.27 percent of GDP in that year, the equivalent of $24 billion presently. As long as there were no phobia about borrowing, a temporary loss of revenue of this magnitude would be tractable.
Government as Stakeholder
If the government becomes the largest shareholder in corporate America, Congress must set a policy for how it uses its influence. It could actively work to promote corporate policies deemed in the public interest, or it could surrender its voting power and simply follow an investment strategy intended to maximize returns.
In recent years, pressure from institutional shareholders like state and local pension funds and university endowments has often been effective in changing corporate policies on various social issues. The most obvious example is the success of the anti-apartheid movement. There are many other instances where firms have stopped anti-union practices, or curtailed environmentally harmful activities, as a result of corporate campaigns, which have usually included a shareholder activism component. Some state and local government pensions have used targeted investments, for example, to favor housing. Others, such as CalPERS, have aggressively intervened to challenge management practices they deem unprofitable, and even to force changes in corporate control.
In principle, the government could exercise considerable influence by voting its shares. However, the Clinton administration has explicitly rejected that course. Politically, there is no way that this Congress would consider a plan to put trust fund money in the stock market that doesn't have some very serious firewalls to block democratic involvement in investment decisions. If the money does go into the market, Congress will undoubtedly require that it only be placed in broad market indexes, which will in turn be managed by an independent board whose members serve long terms (say, 14 years) and cannot be removed, except for malfeasance.
Yet many progressive organizations, most importantly unions, have been trying to gain more control over the social objectives of corporate policy through their ownership stake. The dogma of passive investment could constitute a political setback for these efforts. Even if political reality requires Social Security not to pursue socially conscious investment, individual pension funds and corporate campaigns by unions and their allies have every valid reason to do so.
While trust fund investment in stocks without firewalls is clearly not politically viable at present, it is an open question whether trust fund investment with firewalls is viable. Republicans were quick to denounce the Clinton proposal as "socialism." In principle, it is clearly possible to set up an entity that has a considerable degree of independence from Congress (the Federal Reserve Board is one model). The question is whether the sort of political consensus exists at present to create such an entity. If the independence of this new investment management agency is to withstand the demands of a future congressional majority, then its independence must be greatly valued as an end itself.
But what consensus is possible? For example, would unions accept that the trust fund's money had to be invested in companies that employed prison labor in China or child labor in Indonesia, because this maximized returns? Would environmentalists accept that the fund could not refuse to invest in companies that clear-cut rain forests in Central America or East Asia? Would the Christian right accept that their money was being invested in HMOs that performed abortions or pharmaceutical companies that manufactured abortion-inducing pills? What about tobacco companies, gun companies, and pornographic film distributors? In the absence of such a consensus, effective firewalls will not be created, and trust fund investment in the stock market will not happen.
To date, the closest model, the government's own Thrift Savings Plan for federal employees, has resisted sporadic attempts to impose political agendas on its investment decisions. (Indeed the Federal Reserve itself has an employee pension plan, which presumably doesn't benefit from insider trading.) But Social Security presents a much larger target. Nor does the issue go away if Congress opts for individual accounts. If the accounts are administered through a centralized system, Congress could impose political criteria on the investment decisions of this new agency in exactly the same way it could with trust fund investment. In fact, even if the money were invested in a decentralized manner, there still would be opportunities for political intervention. Any system of mandatory savings would require some regulation. While individuals might be allowed many options, presumably some investments, such as cattle futures and hedge funds managed by Nobel Laureates, would not be deemed appropriate for these accounts. Clearly, political factors could affect this list of acceptable investment options. Moreover, privatized alternatives would incur tens of billions of dollars of needless administrative costs each year.
One of the aspects of trust fund investment that has not received sufficient attention is the potential political effect of the government having a large stake in the profitability of major corporations. Even though the percentage of the market owned by the government would not be that large, a considerable amount of money would be at stake. For example, if shares of Microsoft only grew in value at the same rate as the economy as a whole, the total value of its shares would exceed $800 billion by 2015. This means that a 2.5 percent stake in Microsoft would be worth $20 billion in 2015. Suppose that an antitrust suit against Microsoft might lower the value of its stock by 20 percent. This would cost the trust fund $4 billion.
In reality, this is not a lot of money, particularly in an economy that has a GDP in excess of $19 trillion. Also, the government already has a stake in corporate profits, since it taxes them at about a 35 percent rate. Still, the cost to the government due to a drop in stock prices might be more visible and more easily quantified than a loss of tax revenues. It is a safe bet that the corporate lobbyists of the future will be trumpeting the potential losses to the trust fund in opposing legislation in support of labor, consumers, or the environment. Unless we can be confident that the politicians of the future will not be moved by assertions that raising the minimum wage will lose the trust fund $1 billion due to the drop in McDonalds's stock price, or that restricting tobacco sales to minors will cost the trust fund $2 billion on its holding of tobacco stocks, then the political impact of trust fund investment should be taken very seriously.
To date, the debate on trust fund investment has been narrowly focused on the choice between holding stocks and holding government bonds. In fact, there are many other options that are worth examining. For example, at the simplest level the trust fund could purchase a wider range of government bonds. At present the rate of interest on the infla tion-indexed ten-year bond is 3.8 percent. The interest rate on regular ten-year bonds is approximately 5 percent. Since the inflation rate is projected to be about 2 percent, this means that the real interest rate on the standard ten-year bond is just 3 percent. If the trust fund had the option of selecting among government bonds to maximize returns, it would be able to purchase the inflation-indexed bond and get a somewhat higher yield.
The trust fund could also purchase a variety of government-guaranteed securities. Most of these securities are based on government-guaranteed mortgages, but there are also securities based on government-guaranteed student loans and other types of government-insured debt. These securities pay interest rates up to a full percentage point higher than that available on government bonds. If the trust fund were allowed to invest in these other assets, it could raise the yield on its assets without raising the same sort of political issues and risks raised by stock ownership. Also, given the currently inflated value of the stock market and the low projected growth rate of profits, it is not obvious that the fund would be sacrificing much in the way of returns if it went this route.
It is tragic that the nation's most successful social program is fighting for survival at the end of the twentieth century. The opposition has a powerful ideological and often economic interest in dismantling Social Security. They have managed to undermine confidence in the program through a long, carefully planned campaign of misinformation. Unfortunately, progressives have been very late in responding to this effort. As we try to reverse the tide, it will be necessary to evaluate a range of strategies that we might not have previously considered. While we have to be open to new approaches, we also have to be careful that we don't end up assisting the right's efforts to destroy Social Security.